# what math prerequisites are necessary to understand cutting edge AI algorithms? Are there any study guides for this shit?

what math prerequisites are necessary to understand cutting edge AI algorithms? Are there any study guides for this shit?

1. 1 year ago
Anonymous

Do you know
>statistics
>ODE
>PDE

You're good to go. Only thing you need is to translate the parameter into a program

• 1 year ago
Anonymous

why PDE?

• 1 year ago
Anonymous

AI uses lot of gradient optimization

You don't need to understand algorithms, you need to be able to use a certain implementation. I bet anyone can write Adam optimiser. And it's not AI, it's just some machine learning

• 1 year ago
Anonymous

ML is a subfield of ai

• 1 year ago
Anonymous

Tell that to anyone doing actual ML, eg PhDs/postdocs not uni freshman. lmao

• 1 year ago
Anonymous

What is a vector
what is an array
hardware go bbbrrrrrrr
what is frick around
anything else is cope by bigboy biggurl cathedral midwits like
postdocs maxe garbage toiletpaper no one reads. I daisy chain gpu and make porn for fat dosh.

• 1 year ago
Anonymous

you use code written by postdocs moronic highschooler

• 1 year ago
Anonymous

cool thanx for the free labor Black person slave

• 1 year ago
Anonymous

I get paid a LOT than you

• 1 year ago
Anonymous

kek, imagine the physiognomy of the bugman who reads AI research papers

• 1 year ago
Anonymous

You don't need Ordinary Differential Equations nor Partial Differential Equations to understand how to construct or use a neural network. Those mathematical objects aren't used. In fact, you only has to know what is a function and the four basic operations (+, -, * and /). If you want to go a bit deeper, knowing what is a matrix (a table of values) and what is a "derivative" (the idea, its meaning, not the mathematical expression) and how to minimize a function are enough.

• 1 year ago
Anonymous

And if you really want to know it all, you need basic statistics and probabilities (probability law, expected value, etc.). Nothing crazy.

• 1 year ago
Anonymous

This is also my guess from what I hear from ai people. I guess it depends on the bar for "understanding." Many a data sci researcher, I suspect, are basically monkeys -- don't really know how it works, but can intuit what is useful and create code that appears to do what you want.

2. 1 year ago
Anonymous

This is good shit. Just ask gptchat for a study guide to breach barriers you might find.

https://karpathy.ai/zero-to-hero.html

• 1 year ago
Anonymous

basic math skills and bare minimum knowledge of statistics.

frick off, moron.

• 1 year ago
Anonymous

>frick off, moron.
Are you ok? That site displays the prerequisites and should be a nice start for OP to understand what he needs to study. Asking ChatGPT is a much better choice than flaming buttholes like you who think they are so much better than other people.

But I guess you can't comprehend this, since you are a failed human being.

• 1 year ago
Anonymous

Not OP nor the strange anon that called you moronic but I'll go through that Karpathy lecture tomorrow, looks great thanks anon

3. 1 year ago
Anonymous
• 1 year ago
Anonymous

>meme list
Frick off, midwit.

• 1 year ago
Anonymous

Is there a list like this but for physics and chemistry?

4. 1 year ago
Anonymous

>NO YOU CAN'T JUST DO CALCULATIONS IN MATH IN C YOU HAVE TO CREATE A NODE OBJECT AND AN EDGE OBJECT

5. 1 year ago
Anonymous

How in the name of frick do people learn math? Is there some secret like number sense?

• 1 year ago
Anonymous

You go to uni for it

6. 1 year ago
Anonymous

dr chudner?

• 1 year ago
Anonymous
7. 1 year ago
Anonymous

There can be some slightly more involved math when you're looking at problem specific architectures, but if you just want to understand how a basic ass NN works, you need very little math.

8. 1 year ago
Anonymous
9. 1 year ago
Anonymous

>https://karpathy.ai/zero-to-hero.html
well done!

10. 1 year ago
Anonymous

Thank you Dr. Chud!

11. 1 year ago
Anonymous

To understand cutting-edge AI algorithms, a strong background in linear algebra, calculus, and probability is necessary. Additionally, some knowledge of optimization techniques, such as gradient descent, and basic algorithms, such as dynamic programming and search, would be helpful.

Linear algebra concepts such as matrix operations, eigenvalues, and eigenvectors are used in many AI algorithms, including neural networks. Calculus is used for optimization and training machine learning models. Probability is used for modeling uncertainty and making decisions under uncertainty.

There are many resources available for studying these topics, including online tutorials, textbooks, and MOOCs (massive open online courses). Some popular resources include:

3Blue1Brown (linear algebra, calculus, neural networks)
Coursera (machine learning, deep learning)
edX (probability, optimization)
It's also worth noting that many AI researchers also come from diverse backgrounds, so don't be discouraged if you don't have a strong math background. A lot of the material is learnable and there are many resources to help you learn.

catgpt

12. 1 year ago
Anonymous

>Dr. Chud

13. 1 year ago
Anonymous

Trust me, I'm not a doctor.

14. 1 year ago
Anonymous

>